Transverse aortic constriction induces gut barrier alterations, microbiota remodeling and systemic inflammation

Accumulating evidence suggests that modifications of gut function and microbiota composition might play a pivotal role in the pathophysiology of several cardiovascular diseases, including heart failure (HF). In this study we systematically analysed gut microbiota composition, intestinal barrier integrity, intestinal and serum cytokines and serum endotoxin levels in C57BL/6 mice undergoing pressure overload by transverse aortic constriction (TAC) for 1 and 4 weeks. Compared to sham-operated animals, TAC induced prompt and strong weakening of intestinal barrier integrity, long-lasting decrease of colon anti-inflammatory cytokine levels, significant increases of serum levels of bacterial lipopolysaccharide and proinflammatory cytokines. TAC also exerted effects on microbiota composition, inducing significant differences in bacterial genera inside Actinobacteria, Firmicutes, Proteobacteria and TM7 phyla as shown by 16S rDNA sequencing of fecal samples from TAC or sham mice. These results suggest that gut modifications represent an important element to be considered in the development and progression of cardiac dysfunction in response to TAC and support this animal model as a valuable tool to establish the role and mechanisms of gut-heart crosstalk in HF. Evidence arising in this field might identify new treatment options targeting gut integrity and microbiota components to face adverse cardiac events.

To determine the effects of TAC on gut barrier function, we analyzed circulating levels of FITC-dextran D4000 at different time intervals after oral administration by gavage in sham 1w and TAC 1w mice. In TAC 1w mice, circulating levels of FITC-dextran significantly increased 1 h after oral gavage, and thereafter decreased. In contrast, time-course of FITC-dextran circulating levels was delayed in sham 1w mice compared to TAC 1w, reaching peak concentration 2 h after oral administration, and decreasing thereafter (Fig. 1E). Gut barrier function was still functionally impaired in TAC 4w mice, as shown by differences in circulating levels of FITC-dextran between TAC 4w and sham 4w mice 1 h after gavage (Fig. 1F).
Colon expression levels of anti-inflammatory cytokine interleukin-10 (IL-10) were significantly reduced in TAC 1 w and TAC 4w colon samples compared to respective sham ( Fig. 2A; see Supplementary Fig. 2 for pre-surgery values). These changes were associated to the histological evidence of remarkable inflammatory infiltrate in murine colon samples from TAC 1w mice (Fig. 2B). Consistent with these results, serum levels of  TAC impacts on fecal microbiota composition. Gut barrier integrity is closely linked to gut microbiota composition, and they can be reciprocally affected, especially in response to external pathological causes. Differences in fecal microbiota composition among sham and TAC mice were determined by 16S rDNA sequencing and a Good's coverage index > 99% was obtained at the rarefaction point of 26,396 reads/sample. Comparison of fecal gut microbiota communities among groups revealed significant changes of bacterial genera inside Actinobacteria, Firmicutes, Proteobacteria and TM7 phyla. Specifically, intergroup differences at genus and species levels were analyzed by the linear discriminant analysis (LDA) effect size (LEfSe), identifying the genera Bifidobacterium, Lactobacillus, Turicibacter, unclassified genus (u.g.) of RF32 and u.g. of F16 as characteristic of TAC mice, whereas the genus Oscillospira was significantly less abundant in TAC mice compared to sham at specific time windows (Fig. 3A). SPINGO high-resolution approach was used to obtain bacterial species assignment of key genera with significant differences among groups (Fig. 3B). After TAC, a significant increase in OTUs resembling Bifidobacterium pseudolongum, Turicibacter sanguinis, Lactobacillus frumenti, and an unclassified species belonging to Proteobacteria phylum (order RF32) was detected, along with decrease of species within Oscillospira genus (Pseudoflavonifractor capillosus and Flavonifractor plautii, Fig. 3B).
In order to predict functional effects of gut microbiota alterations induced by TAC, we performed a Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUST) analysis. This

Discussion
In the present study we demonstrate that the murine model of TAC induces prompt and strong weakening of intestinal barrier integrity, decrease of colon anti-inflammatory cytokine levels, increase of serum levels of LPS and proinflammatory cytokines, and significant differences in fecal bacterial genera inside Actinobacteria, Firmicutes, Proteobacteria and TM7 phyla. Our findings clearly support this murine model as a valuable tool to establish the importance of gut barrier function and microbiota composition in HF. Several studies have demonstrated that a decrease in gut perfusion structurally and functionally affects intestinal barrier integrity 19 . In patients with HF, both alteration of villi shape in intestinal mucosa resulting in intramucosal acidosis 20,21 and intestinal wall thickness with edema have been demonstrated 22 . Moreover, even if with multiple discrepancies, alterations in gut microbial communities have been reported in patients with HF [22][23][24][25][26][27][28][29] . Our experiments showed that TAC was associated with a prompt reduction in abdominal aortic blood flow and intestinal hypoperfusion, accompanied by a rapid impairing of gut barrier structure as demonstrated by reduction of zonula occludens-1 and occludin levels and increased FITC-dextran translocation across intestinal epithelium into blood. Moreover, we concurrently found a significant increase of LPS and proinflammatory cytokines in systemic blood circulation of TAC mice compared to sham. Of note, together with remarkable inflammatory infiltrate persistence and high levels of inflammatory markers, gut barrier was still functionally compromised in TAC 4w mice, as revealed by FITC-dextran in vivo permeability assay, despite a recovery of tight junctions mRNA and proteins levels at this time point. We cannot exclude that at later time points gut barrier function could be restored.
These dramatic alterations of gut barrier due to hypoperfusion induced by TAC were associated with remarkable changes in gut microbiota composition. Based on LEfSE results, taxa correlated to dysbiosis and colonic inflammation, such as F16, significantly increased after TAC [30][31][32] . Furthermore, we identified unclassified genus of RF32 (Proteobacteria phylum) as a potential microbial biomarker and source of LPS in TAC operated mice,  Figure 3. Gut microbiota composition after sham or TAC surgery in mice. Gut microbiota differences based on 16S rDNA sequencing at genus (A) and species (B) taxonomic levels were identified using linear discriminant analysis (LDA) combined with effect size (LEfSe) algorithm. In each panel, LDA scores (left) and relative abundance (right) of key phylotypes discriminating sham and TAC bacterial communities are reported (sham n = 8; TAC n = 9). Statistical significances were assessed using LEfSe analysis with alpha values of 0.05 for both Kruskal-Wallis and pairwise Wilcoxon tests and a cutoff value of LDA score (log10) above 2.0 (*p < 0.05 and **p < 0.01 vs. correspondent sham). www.nature.com/scientificreports/ possibly contributing to colonic inflammation. Moreover, in TAC 4w mice gut microbiota was characterized by an increase in genera Turicibacter and Lactobacillus. The increase of these lactate-producing bacteria (namely T. sanguinis and L. frumenti, according to species-level SPINGO classification) concurred with the depletion of butyrate-producing bacteria (genus Oscillospira, taxonomically classified as P. capillosus and F. plautii species after SPINGO procedure) in TAC mice, and these results were also corroborated by PICRUST analysis. Recent studies have implicated increases in genus Lactobacillus in pathophysiology of cardiovascular diseases, even if conflicting results have been reported in HF. Increases in lactate-producing Lactobacillus have been demonstrated in elderly patients with HF and in animal models of hypertension 24,33 . Consistently, ST-segment elevation myocardial infarction (STEMI) patients were characterized by increased circulating levels of intestinal Lactobacilli, associated with systemic inflammation and adverse cardiovascular events 33 . On the contrary, ferulic acid administration, which improves cardiac function in TAC mice, has been shown to increase intestinal Lactobacillus 34 . Moreover, a reduction of Lactobacillus has been reported in gut microbiota of rats with isoproterenol-induced HF 35 . Thus, further investigations will be necessary to clarify the role of Lactobacillus phylotypes in the pathophysiology of HF, ideally with species-level approaches. A significant depletion of genus Oscillospira, an under-investigated bacterium usually considered a biomarker of intestinal and host wellness, was also identified in TAC mice. Presence of this bacterium has been associated to leanness, resulting depleted in obese patients and reduced in diseases that involve inflammation, and these associations have been inferred to the putative capacity of Oscillospira species to produce butyrate 36 . Butyrateproducing bacteria are considered relevant colonizers of the gastrointestinal tract with known anti-inflammatory effects and a prominent role in maintaining intestinal barrier integrity 37,38 . Consistent with our findings in TAC mice, butyrate-producing bacteria such as Faecalibacterium and Blautia genera, were also depleted in other animal models of hypertension 39 and in studies involving patients with HF [24][25][26] . Thus, the unbalance between butyrate and lactate producing bacteria might represent a possible mark of gut microbiota adaptations to the new environment induced by intestinal hypoperfusion in presence of HF. Overall, we propose that gut hypoperfusion induced by TAC structurally and functionally affects intestinal barrier with an effect on the balance of gut microbiota composition. TAC-induced gut dysbiosis and enhanced gut permeability may in turn affect systemic inflammation possibly contributing to cardiac dysfunction.

Conclusions
Alterations of gut structure/function and dysbiosis may represent important elements to be considered in the development or progression of cardiac dysfunction in response to pressure overload induced by TAC. Whether restoration of gut function and microbiota composition might exert a beneficial effect on cardiac remodeling and dysfunction is still unknown and will require further investigations. However, our findings clearly support the murine model of TAC as a valuable tool to establish the importance of gut barrier function and microbiota composition in HF, suggesting novel important avenues of research in this field, including administration of single or defined cocktails of bacterial species to counteract alterations in gut barrier integrity during HF. Experimental animals. C57BL/6 (Charles River Laboratories) mice of either sex (8-week-old) were included in the study and were maintained under identical conditions of temperature (21 ± 1 °C), humidity (60 ± 5%), and light/dark cycle of 12 h and had free access to water and normal mouse chow diet. One month before either sham or TAC surgery, mice generated by different mothers were placed in cages (groups of 2-3 mice/cage). After surgery, mice were kept in single cages until study termination.

Methods
Mouse model of pressure overload-induced cardiac hypertrophy and heart failure. Pressure overload was induced in adult C57BL/6 mice by TAC as previously described 40 . Mice were anesthetized by administration of 5% sevoflurane and 95% O 2 and a suture was surgically placed across the aortic arch between the left common carotid artery and the innominate artery. Another group of animals underwent a left thoracotomy without aortic constriction (sham). Mice from sham and TAC groups were sacrificed 1 week (1w) and four weeks (4w) after surgery to perform molecular analyses. Only TAC animals with systolic pressure gradients > 40 mmHg measured by Doppler echocardiography were included in the study (see below). Cardiovascular ultrasound and Doppler. Cardiac function was non-invasively monitored by transthoracic echocardiography, using the Vevo 2100 high-resolution imaging system (Visual-Sonics, Toronto, ON, Canada) before, 1w and 4w after surgery as previously described 41 . Briefly, mice were anesthetized by an intraperitoneal injection of 0.1 ml/kg of mixture of 50% tiletamine and 50% zolazepam (Zoletil 100) and echocardiography was performed. Echocardiographic measurements were obtained from grayscale M-mode images at the mid-papillary level in the parasternal short-axis. Conventional measurements of the left ventricle (LV) included: left ventricular end-diastolic diameter (LVEDd), left ventricular end-systolic diameter (LVESd), interventricular septum end-diastolic diameter (IVSd), posterior wall end-diastolic diameter (PWd), heart rate (HR), % fractional shortening (FS %) and % ejection fraction (EF%). Pressure gradients across transverse aorta induced by Scientific Reports | (2021) 11:7404 | https://doi.org/10.1038/s41598-021-86651-y www.nature.com/scientificreports/ TAC were evaluated using pulsed wave (PW) Doppler analysis 1w and 4w after surgery. Abdominal aortic flow was non-invasively estimated by PW Doppler before, 1, 3, 7, and 28 days after TAC or sham surgery.

Microbiota sequencing and data analysis.
To study fecal microbiota composition of sham and TAC mice, 16S rRNA gene sequences were obtained and pre-processed as previously described 42 . Resulting data were analyzed using Quantitative Insights Into Microbial Ecology (QIIME, version 1.9.1) 43 . Microbiota reads were collapsed into operational taxonomic units (OTUs) using a closed reference-based OTU picking method against Greengenes 16S gene database (GG, may 2013 version) 44  neutral buffered formalin and embedded in paraffin wax; sections were cut and stained with haematoxylin and eosin (H&E; Carlo Erba, Italy). Sections were examined and scored using a 0-to-4 scale as previously described 42 .
Histological scoring system was as follows: (a) the severity of inflammatory cell infiltration was evaluated by percentage of leukocyte density in lamina propria area and estimated in a high-power field (HPF) representative of the section (0 for no signs of inflammation, 1 for minimal < 10%, 2 for mild 10-25%, 3 for moderate 26-50%, 4 for marked > 51% with dense infiltrate); (b) the extent of the inflammation was estimated as expansion of leukocyte infiltration (0 for none, 1 for mucosal, 2 for mucosal and submucosal and 3 for mucosal, submucosal and transmural level); (c) the presence of fibrosis (0 for none, 1 if present). For immunohistochemistry, 4-µm-thick sections were processed with the MACH1 Universal HRP Polymer Detection Kit (Biocare Medical LLC, Concord, CA). Colon tissues were mounted on positively charged glass slides (Bio-Optica, Milan). Antigen retrieval pretreatments were performed using a heat-induced epitope retrieval (HIER) citrate buffer pH6.0 (Bio-Optica, Milan, Italy) for 20 min at 98 °C; peroxide block was applied for 15 min at room temperature, and then the sections were incubated for 30 min with background sniper (Biocare Medical LLC). The primary antibodies were diluted in phosphate-buffered saline (PBS) and incubated overnight at 4 °C. Horseradish peroxidase (HRP) polymer was added for 30 min at room temperature. After every step, the sections were washed in 0.01 M PBS (pH 7.2-7.4). The reaction was revealed by using 3,30-diaminobenzidine www.nature.com/scientificreports/ Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.